Re: [TRNSYS-users] Request

[Knut Erik Enerstvedt <knut.erik.enerstvedt@gmail.com>]

Dear Tadashi Takakura,

I have been working on greenhouse models in TRNSYS for a while in a long-term project. 
Most of the suggestions from forerunners I followed you can find in the archive by following 
the links provided by Matt. In addition, I wish to add some comments:

I do indeed use two zones in my Type 56 greenhouse model, as you suggest. The zones are 
separated by a thermal screen/horizontal shade modelled by a window with no glazing (g-value=1). 
Then you can have an internal shading device on the "window" controlled by an input variable
(shades are usually controlled based on temperature, radiation and humidity level). I think you are 
fortunate to have TRNSYS 17, as I understand it is better suited than TRNSYS 16 for modelling 
greenhouses because of a more accurate way of treating solar radiation passing through a zone.
My model is based on TRNSYS 16.

You will also have to model the air exchange between the two zones. As mentioned in the archive, 
there are some problems related to CONTAM for air flows through a large horizontal opening. I use a 
model equation found in this thesis to calculate the air flow:

Hendrik Feije de Zwart, 1996: "Analyzing energy-saving potensials in a greenhouse cultivation
using a simulation model". PhD. Dissertation, Landbouwuniversiteit, Wageningen.

However, the air flow is too large for TRNSYS to converge when the screen is fully open, so I
have to limit it.   

Except for the floor, which is defined as a boundary wall, I have no other walls in my greenhouse.
All external surfaces have been defined as "additional windows".

As constant infiltration rate (except when opening windows), I use 1 ACH (by advise from

a greenhouse expert). When opening windows to release heat and humidity, I add this infiltration
to the zone which contains the windows (upper in my case). The infiltration rate is calculated
by an equation from the above mentioned thesis.

The plants make a substantial contribution to the humidity level in a greenhouse. I have created
my own Type for modelling the plants, based on the above mentioned PhD thesis. If you know
the approximate water consumption of the plants in the greenhouse you are modelling, you can
always use this to estimate water vapour flow rate into greenhouse air. But then you will of course
miss most of the transient behaviour due to changes in radiation and temperature. I connect water
vapour flow rate from the plants to Type 56 as a humidity gain. There is also a thermal interaction
between the plants and Type 56, as the canopy temperature normally lies 0.5-1 deg below air 
temperature when evaporization is taking place and the plants are not exposed to direct sunlight.

I hope this information has been helpful. Please do not hesitate to contact me if you need further
details. I am also looking forward to possibly sharing experiences about greenhouse modelling.


Best regards,
Knut Erik Enerstvedt
   

>Dear Tadashi Takakura,

>I imagine the difficult task for this project would be accurately
>accounting for the humidity in the zones. I have never used the feature,
>but you could perhaps look into the detailed humidity mode for Type56
>building. Again, with the new TRNSYS 3D, you can create an upper and lower
>zone(or ‘airnode’ as the refer to it in TRNBuild) with the ‘Virtual
>Surface’. There is more information in the tutorial and documentation.

>You can also look at the archives from the TRNSYS Users mailing list. My
>colleague, David, provided much useful information. Greenhouses have been
>discussed a few times. The following are the links to the archives.

>https://www-old.cae.wisc.edu/pipermail/trnsys-users/2009/004091.html
>https://www-old.cae.wisc.edu/pipermail/trnsys-users/

>Best regards,

>Matt Duffy



>----- Original Message -----
>From: "Tadashi" <takakura>
>Sent: Sun, August 22, 2010 20:05
>Subject:[TRNSYS-users] Request


>    We just started to make greenhouse models using TRNSYS17.  Probably we
>will select Type 56 building and have two zones, one for upper air space and
>he other zone with plant canopy.  Combination with CONTAM would be the next
>step.

 >    Any suggestions and comments which can be given by forerunners would be
>very much appreciated.

>    Best regards,

 >    Tadashi Takakura